bluewave

We could get another big track jump by 0z or 12z. The ICON just came pretty far west at 18z. Maybe it will take another run or two for the models to converge.

This will be the first storm for the upgraded HRRR once we get within its effective range,

The ridge verifying stronger to our south and east than longer range model guidance has been our new normal in recent years. There have been so many week 2 forecasts that suddenly correct stronger within the day 7-10 range. It would be nice if we had bias corrected model maps to deal with this. Maybe this will be the next step in long range model forecasting. The exception to the rule is strong -AO and -NAO blocking. https://www.ecmwf.int/en/about/media-centre/science-blog/2019/will-machine-learning-replace-conventional-weather How could machine learning be used in numerical weather prediction? There are a large number of possible application areas for machine learning throughout the workflow of numerical weather prediction. Here, the “system” could be a tool to pick and bias-correct observations that are used within data assimilation; a tool to project real-world observations into a representation of variables as they are used within models; or a part of the numerical forecast model – such as the radiation scheme which predicts the heating of the atmosphere given surface temperature, cloud patterns and incoming radiation from the sun. It could also be a tool that is used in post-processing of the output of a weather forecast model, for example to detect extreme weather events in forecasts or to bias-correct forecasts for a specific location.

A different day and another set of solutions.

It did very well from Hurricane Sandy in 2012 to Nemo in February 2013. But really stumbled with the January 2015 event. It managed to do very well with Joaquin in 2015. Too suppressed with the January 2016 blizzard. Also too close to Houston with Laura. So maybe the amazing job it did with Sandy just set expectations too high. That will stand as the forecast of the century in my book.

I guess model mayhem is the end result when the pattern tries to shift from more Niño-like to more Niña-like in the same month.

I can’t remember the last time the Euro jumped around this much from run to run. Maybe one of the recent upgrades threw the model off?

The minimum temperature record was more impressive than the maximum. The max beat the DJF record by 1°. But the min beat the previous record by 5°. Time Series Summary for Caribou Area, ME (ThreadEx) - Month of Dec Click column heading to sort ascending, click again to sort descending. Rank Year Highest Min Temperature Missing Count 1 2020 50 30 2 1964 45 0 3 2003 43 0 - 1939 43 6 5 2011 40 0 - 1951 40 0 Time Series Summary for Caribou Area, ME (ThreadEx) - Month of Jan Click column heading to sort ascending, click again to sort descending. Rank Year Highest Min Temperature Missing Count 1 1950 41 0 2 1986 40 0 3 1979 38 0 - 1956 38 0 4 2008 37 0 - 2007 37 0 5 2017 36 0 Time Series Summary for Caribou Area, ME (ThreadEx) - Month of Feb Click column heading to sort ascending, click again to sort descending. Rank Year Highest Min Temperature Missing Count 1 1994 42 0 2 1970 41 0 3 1981 39 0 4 1988 38 0 5 1996 37 0 Time Series Summary for Caribou Area, ME (ThreadEx) - Month of Dec Click column heading to sort ascending, click again to sort descending. Rank Year Highest Max Temperature Missing Count 1 2020 60 30 2 1950 58 0 3 2010 57 0 4 2001 56 0 - 1969 56 0 5 2000 54 0 - 1982 54 0 - 1973 54 0 - 1966 54 0 Time Series Summary for Caribou Area, ME (ThreadEx) - Month of Jan Click column heading to sort ascending, click again to sort descending. Rank Year Highest Max Temperature Missing Count 1 2013 53 0 - 1995 53 0 2 1986 52 0 3 2018 51 0 - 2005 51 0 - 1996 51 0 - 1983 51 0 - 1950 51 0 4 2006 50 0 5 2008 48 0 - 1999 48 0 - 1945 48 0 Time Series Summary for Caribou Area, ME (ThreadEx) - Month of Feb Click column heading to sort ascending, click again to sort descending. Rank Year Highest Max Temperature Missing Count 1 1994 59 0 2 1990 52 0 - 1981 52 0 3 2018 51 0 4 2017 50 0 - 2016 50 0 5 2000 49 0 - 1988 49 0 - 1976 49 0

The persistent ridge east of New England continues to beat expectations.

All models have their individual biases. That why I pointed out the other day that the Euro was leaving too much energy back to the SW. So it’s now more phased for the 5th instead of squashing the low off the SE Coast. New run Old run

The parallel GFS is actually closer to the Euro. But it’s not quite as amped yet.

It’s a variation on the same theme. Last December 2nd had the low track just south of Long Island also. But there was more cold available to start last December than this year near the coast. It’s interesting that we are seeing a similar track almost a year to the day.

Thanks. Updated for November 2020. 11....2020...EWR....4...NYC...1....LGA....1...JFK...4...BDR...5....ISP....4

Even the EPS drops the cold into Montana in about 8-10 days. But the GEPS and GEFS dig a little more into the Rockies. Montana and surrounding areas have really been the focus of what little cold departures there have been in recent years.

First 60° temperature on record today in Caribou during meteorological winter.

These differences in handling the short term storm details really magnifies the ensemble spread by day 10. That’s why I want to see how each individual storm verifies before looking past day 8-10. The EPS, GEFS, and GEPS all have varying solutions day 8-10.

The 12z Euro finally ejects enough energy from the SW for a more phased coastal rain and wind event on the 5th. 12z today 12z yesterday

Updated for November 2020. 11....2020...EWR....4...NYC...1....LGA....1...JFK...4...BDR...5....ISP....4

I was pointing out back in the October and early November threads how the atmosphere wasn’t showing the canonical La Niña patterns. We have been experiencing these more Niño-like +AAM spikes since the fall. So I mentioned that we may not follow a typical La Niña November. This is much like the last few years with mismatched Niño and Niño-like influences at the same time. This is due to all the competing record equatorial and off equatorial SST warmth from the Indian Ocean into the Pacific.

Those long range strat forecasts for change are very uncertain given how strong the PV has been.

The OP GFS longer range has had less of a cold bias than the ensembles over SE Canada. Sometimes, the lower resolution ensembles smooth out the ridge near the near the Northeast too much. Where the OP runs don’t.

It only took NYC 5 years to beat 2015 for warmest November. It took 14 years for 2015 to surpass 2001. And 2001 over 20 years to go ahead of 1979. Time Series Summary for NY CITY CENTRAL PARK, NY - Month of Nov Click column heading to sort ascending, click again to sort descending. Rank Year Mean Avg Temperature Missing Count 1 2020 53.0 0 2 2015 52.8 0 3 2001 52.7 0 4 1979 52.5 0 5 1948 52.4 0 6 1975 52.3 0 7 2011 51.9 0 - 2006 51.9 0 - 1994 51.9 0 - 1931 51.9 0 8 1902 51.4 0 9 2009 51.1 0 10 1999 50.8 0

The Euro is really struggling with the split flow. Notice how much different the GFS and CMC are at day 10. Remember, the errors at day 7-10 often build into the day 11-15.

People will probably replant smaller ornamentals that have less chance of damaging their homes in our more extreme climate.

Longer range, we need to watch for the ridge east of New England verifying stronger than forecast. This has been the pattern for a while now. Today is just the most recent case. 7 day Forecast Verification